Skip to main content
Glama
README.md
# Archify MCP Server 🏗️

> MCP Server chuyên phân tích dự án và đề xuất công nghệ phù hợp

## 🎯 Features

| Tool | Description |
|------|-------------|
| `analyze_project` | Phân tích chuyên sâu ý tưởng dự án |
| `recommend_stack` | Đề xuất tech stack phù hợp (40+ technologies) |
| `suggest_features` | Đề xuất tính năng theo MoSCoW priority |
| `compare_tech` | So sánh công nghệ side-by-side |
| `design_architecture` | Thiết kế kiến trúc với Mermaid diagrams |
| `estimate_project` | Ước lượng timeline, team, và chi phí |
| `generate_roadmap` | Tạo roadmap với Gantt chart |

## 🚀 Quick Start

```bash
# Install dependencies
npm install

# Build
npm run build

# Run
npm start
```

## ⚙️ Configuration

Add to your MCP client config (e.g., Claude Desktop):

```json
{
  "mcpServers": {
    "archify": {
      "command": "node",
      "args": ["/Users/tai/Desktop/Archify/dist/index.js"]
    }
  }
}
```

Or for development:

```json
{
  "mcpServers": {
    "archify": {
      "command": "npx",
      "args": ["tsx", "/Users/tai/Desktop/Archify/src/index.ts"]
    }
  }
}
```

## 📊 Supported Project Types

`ecommerce` • `saas` • `social_network` • `marketplace` • `content_platform` • `fintech` • `healthtech` • `edtech` • `iot` • `ai_ml` • `gaming` • `enterprise` • `mobile_app` • `api_service` • `devtool`

## 🛠️ Technology Knowledge Base

40+ technologies scored across 8 criteria:
- Performance, Scalability, Developer Experience
- Ecosystem, Security, Cost Efficiency
- Documentation, Community Support

## 📐 Architecture Patterns

8 patterns: Monolith, Modular Monolith, Microservices, Serverless, Event-Driven, JAMstack, Clean Architecture, BFF

TDQS

A3.5/5.0

Scored across 26 tools

Disambiguation3/5

Most tools map to distinct planning domains, but smart_recommend, recommend_stack, compare_tech, analyze_compatibility, and recommend_dual_stack all touch stack selection and could be confused. Tool descriptions clarify their specific purposes, but an agent still faces meaningful overlap when choosing among them.

Naming Consistency5/5

All 26 tools consistently follow a verb_noun snake_case pattern such as plan_testing, design_database, and estimate_project. The verbs vary semantically, but the naming convention is uniform and predictable across the entire toolset.

Tool Count3/5

26 tools is on the heavy side, just above the 16-25 borderline range defined in the rubric. Each tool covers a distinct planning artifact and full_analysis aggregates them, but the set could be tightened by consolidating several overlapping recommendation and estimation tools.

Completeness5/5

The toolset covers the full architecture-planning lifecycle from project analysis, features, stack selection, architecture, database, security, and cost to estimation, roadmap, boilerplate, testing, DevOps, scalability, team, integrations, and monetization. There are no obvious dead ends, and full_analysis acts as a closed-loop pipeline for complete project assessment.

Maintenance

ActivityInactive
ResponsivenessNo issues